Whenever load-following is necessary the best way (marginal cost, emissions...) to provide it is preferred, and a nuclear is better, on those account, than a gas turbine. Therefore if a gas turbine takes the load, nuclear could not do so (or there is some widespread and long-going sabotage?).
All mechanisms used to tune power are pauses-inducing (depending upon the fuel state...), they cannot be used at will. If I'm wrong please explicitly write here that a "the (thermal) power generated by an existing and active industrial French nuclear reactor can always be freely modulated, without any limit nor any need to sometimes abstain from doing so for a while".
Load following isn't about a plant being started or stopped, it's an automatic mechanism that is either handled without any intervention (primary reserve : the plant's turbine monitors the frequency and adapts over 30s) or with interventions from the grid operator only (secondary reserve: RTE's systems have the hand on the providers' facilities). Every plant that participates in load following does it through the same mechanism, nuclear or gas doesn't matter.
What you describe is merit order, and it's not happening at the same operating level.
> Therefore if a gas turbine takes the load, nuclear could not do so
We do have examples of nuclear being used this way. It is likely that gas plants are also used this way, since every power provider is required to participate, and some french providers don't have NPPs.
> All mechanisms used to tune power are pauses-inducing (depending upon the fuel state...), they cannot be used at will.
Not sure what you mean by that. Like gas plants, nuclear plants work by heating water. This means that this water buffers some energy, and that buffer can be used for fast but small power variations, with minimal impact on the pipe system.
As for large variations in power, such as the 100% -> 20% change you quoted, they are scheduled ahead of time, and you can find technical explanations on how operators do that.
As you noted, there are obviously bounds to variations. As you said, they can't be used at will. One thing you seem to neglect, though, is that flexibility requirements are within these bounds. That means that power variations required can be serviced while staying within the limits of the system.
This means that what you're asking for: "active industrial French nuclear reactor can always be freely modulated" would be unnecessary gold-plating.
> flexibility requirements are within these bounds. That means that power variations required can be serviced while staying within the limits of the system.
This is, indeed, central here.
It seems to me that, if nuclear can do all necessary follow-up (the fine, low-latency, the part of it which is at best not-well-planned and often absolutely not planned for), it should do so, as it is cheaper and emits less than any fossil-fuel-burning equipment.
I can only see two reasons for this:
- the price-calculation method (marginal cost...) used in Europe offers way more benefits by always producing the last kWh thanks to fossil fuel
- letting a fleet of nuclear reactors take on all necessary load-follow has some unwanted long-term effect (costs, maintenance, fuel state...)
Frequency controlling a grid (e.g. increasing or decreasing plant output by a few %s) is important for grid stability first and foremost. Grid costs are less important, and the process is not overseen by the producer, who bears the costs, but by the grid manager. Because of that, every power provider connected to the grid, is required to be able to provide some % of flexibility on its non-renewable fleet, including the providers who only operate fossil plants. For redundancy reasons and ease of implementation, this flexibility is split between as many plants as possible for each operator.
The step at which you can prioritize less costly means of production is scheduled ahead of time. This is when you typically need larger variations, such as moving a plant's output from 100% to 20%. In this step, merit order is used, and, to my knowledge, nuclear has priority over gas in France.
> Frequency controlling a grid (e.g. increasing or decreasing plant output by a few %s) is important for grid stability first and foremost. Grid costs are less important
> redundancy reasons and ease of implementation, this flexibility is split between as many plants as possible for each operator.
This is the core of the argument. My point is that if the nuclear fleet was flexible enough to provide 100% of the follow-up it would do so (because it emits less and costs less), and therefore production snapshots would rarely show significant production from flamme plants (burning fossil fuel), which would only be significant during peak consumption. In other words thanks to such a sufficient flexibility the merit-order could be sound at any moment. The reality is that those 'fossil' plants very often (nearly constantly) generate a fair fraction of the gridpower (they aren't in minimal production "ready to warm-start" mode, they generate in a useful way).
> The step at which you can prioritize less costly means of production is scheduled ahead of time. This is when you typically need larger variations, such as moving a plant's output from 100% to 20%. In this step, merit order is used, and, to my knowledge, nuclear has priority over gas in France.
Yes, and it shows that the nuclear fleet can always (bar any incident) tackle a rather large scheduled (in hours) modulation, which is a totally different challenge than "realtime" follow-up.
All mechanisms used to tune power are pauses-inducing (depending upon the fuel state...), they cannot be used at will. If I'm wrong please explicitly write here that a "the (thermal) power generated by an existing and active industrial French nuclear reactor can always be freely modulated, without any limit nor any need to sometimes abstain from doing so for a while".